assess-observability
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external, untrusted telemetry data while possessing the capability to suggest or execute code changes.
- Ingestion points: The skill ingests data from production or staging telemetry queries, resource dumps, and system probes as specified in SKILL.md and readiness-bar.md.
- Boundary markers: The instructions in SKILL.md do not define specific delimiters or instructions for the agent to ignore potentially malicious content embedded within the ingested telemetry data.
- Capability inventory: The skill is authorized to use tools mentioned in SKILL.md that modify repository files and configurations, such as frame-change, amend-feature, and /define-system-doc.
- Sanitization: There is no evidence of automated sanitization or filtering of the ingested data in SKILL.md or readiness-bar.md beyond manual instructions to check for PII.
Audit Metadata